1

Chapter 1

Introduction

2

1.1 CHEMICAL SCALE NEUROBIOLOGY

One of the most challenging topics facing modern biology is understanding the

complexities of neurobiology. The brain uses an intricate network of neurons to

communicate information. The process of propagating information from one neuron to

the next, termed synaptic transmission, is depicted in Figure 1.1. In this process, the

axon of a pre-synaptic neuron sends an electrical signal, or action potential, to release a

small molecule neurotransmitter (ligand) across a small synaptic cleft. A membrane

receptor, or ligand gated ion channel, located on the dendrite of the post-synaptic neuron

binds the ligand and undergoes a conformational change to allow the passage of ions

through the otherwise impermeable cell membrane. Thus, the electrical signal of the

action potential is converted to a chemical signal through the release of a

neurotransmitter that is then converted to an electrical signal through the gating of a

ligand gated ion channel (LGIC).

Figure 1.1. Synaptic Transmission. Neurotransmitters in an axon are released from vesicles across the synaptic cleft and bind to receptors on post-synaptic dendrites. Ligand gated ion channels are one type of receptor that binds the neurotransmitter, undergoing a conformational change to allow the passage of ions through the otherwise impermeable cell membrane.

Synapse Neurotransmitter

Receptor/Ion Channel

Ligand-Gated Ion Channel

3

Ligand gated ion channels (LGIC) are transmembrane proteins implicated in

Alzheimer’s disease, Schizophrenia, drug addiction, and learning and memory.1,2 The

ability of neurotransmitters to bind and induce a conformational change in these dynamic

proteins is not fully understood. A number of studies have identified key interactions that

lead to binding of small molecules at the agonist-binding site of LGICs. High-resolution

structural data on neuroreceptors are only just becoming available,3-8 yet functional data

are still needed to further understand the binding and subsequent conformational changes

that occur during channel gating.

The primary focus of the present work is to gain a chemical scale understanding of

the ligand-receptor binding determinants of LGICs. In particular, these studies explore

drug-receptor interactions at the nicotinic acetylcholine receptor (nAChR), the most

extensively studied members of the Cys-loop family of LGICs, which include γ-

aminobutyric, glycine, and serotonin receptors. Remarkably, several ligands are known

to bind to the same region of the protein while eliciting different responses in protein

activity, raising interesting chemical recognition questions. In addition, nAChRs are

interesting therapeutic targets, and natural products, nicotine, epibatidine, and cytisine,

serve as lead compounds for drug discovery.2 For example, cytisine served as the lead

compound for Pfizer’s smoking cessation drug candidate, Varenicline, that targets

nAChRs.9 Therefore, chemical scale insights into drug-receptor interactions at the

nAChR are interesting both from a chemical recognition perspective and from a drug

discovery perspective.2

4

1.2 UNNATURAL AMINO ACIDS

In Vivo Nonsense Suppression

The present study performs chemical scale investigations of nAChR agonist activity

though incorporation of unnatural amino acids. The ability to incorporate an unnatural

amino acid site specifically into proteins is achieved through a method termed in vivo

nonsense suppression.10,11 This powerful tool has enabled successful structure-function

studies of several ion channels including nicotinic acetylcholine receptors,10 Shaker

potassium channels, KV2.1 potassium channels,12 5HT3 serotonin receptors,13 and more

recently GABAC receptors.14 The present study applies this well-established method of

incorporating unnatural amino acids to probe drug-receptor interactions at the nAChR.

This method is outlined in Figure 1.2. The mRNA encoding the ion channel is

mutated to contain a UAG Amber stop codon. An orthogonal suppressor tRNACUA

containing a chemically ligated unnatural amino acid recognizes the UAG codon. The

endogenous translational machinery incorporates the unnatural amino acid into the

protein at the site of interest. The incorporation of unnatural amino acids into ion

channels utilizes Xenopus oocytes10 or mammalian CHO cells15 as the translational

machinery host. As shown in Figure 1.2B, the synthesized mRNA (1) and orthogonal

Tetrahymena thermophila tRNACUA (2), containing the stop codon and unnatural amino

acid respectively, are introduced into the cell (3). The translational machinery of the cell

then produces a full-length protein with the unnatural amino acid at the site of interest

(4). This powerful method enables the translation of functional ion channels embedded

in the membrane of the cell. The functional characteristics of the mutated membrane

proteins are monitored through sensitive two-electrode voltage clamp assays (5).

5

Figure 1.2. In Vivo Nonsense Suppression. A) Nonsense Suppression. mRNA containing a stop codon (UAG) is recognized by an orthogonal suppressor tRNACUA that contains a chemically acylated unnatural amino acid (shown in red). The translation machinery incorporates the unnatural amino acid into the protein to produce full-length protein containing a single unnatural amino acid at the site of interest. B) In Vivo Nonsense Suppression in Xenopus Oocytes. mRNA and tRNA are injected into Xenopus oocytes, protein translation occurs, and protein function is monitored using high throughput electrophysiology.

A

B

6

The chemical acylation of an unnatural amino acid to in vitro transcribed orthogonal

tRNA is shown in Figure 1.3. First, amino acids must be prepared by protecting the α-

amine group with a photo-labile or I2 cleavable protecting group to prevent

destabilization of the free amine. It is often unnecessary to protect the α-hydroxy group

of hydroxy acid analogues.16 The carboxylic acid is then activated to react with dCA as a

cyanomethyl ester.10,16 The α-N-protected cyanomethyl ester or the α-hydroxy

cyanomethyl ester is coupled to dCA dinucleotide and then ligated to a truncated 74 base

tRNACUA with T4 RNA ligase to yield the amino-acylated 76 base tRNACUA (aa-tRNA).

Immediately prior to injection into Xenopus oocytes, the α-N-protecting NVOC group or

4PO group is deprotected with 350 nm light or I2, respectively.

Figure 1.3. Preparation of Aminoacyl tRNA.17

7

History of In Vivo Nonsense Suppression

The methodology of incorporating unnatural amino acids in biological systems was

developed by Peter Schultz in 1989.18 The Schultz group combined several experimental

observations in the field of nonsense suppression to develop this method. They utilized

prior knowledge of the ability of amber suppressor tRNAs to recognize amber nonsense

TAG stop codons and block translation suppression. In addition, studies demonstrating

the ability of tRNA recognition to be independent of amino acid identity were utilized in

combination with knowledge of the ability of the ribosome to incorporate a broad range

of amino acid side chains. Precedent for the chemical strategy of aminoacylating

suppressor tRNAs with amino acids was set by the work of Hecht and Brenner. The

combination of these observations led to the first example of site-specific incorporation

of unnatural amino acids into proteins.18

More recently, this method was adapted at Caltech and optimized for in vivo nonsense

suppression of unnatural amino acids to probe ion channel structure-function

relationships in Xenopus oocytes10,11 and mammalian cells.15 A limitation of this method

is the small quantity of protein produced in cells. The translational host can only produce

as much protein as the amount of aminoacylated tRNA, a stoichiometric reagent, present

in the cell. Fortunately, studies of LGICs utilize a highly sensitive electrophysiology

assay where protein amounts produced from nonsense suppression experiments are

sufficient to monitor channel function. To date, Dougherty and co-workers have

incorporated 100 unnatural amino acids into 20 different proteins at 140 different sites

using this method. Structures of many of these successfully incorporated unnatural

amino acids are shown in Figure 1.4.

8

Figure 1.4. Unnatural Amino Acids Incorporated into Ion Channels Using In vivo Nonsense Suppression.

Alternate Methods of Incorporating Unnatural Amino Acids

While the present work utilizes in vivo nonsense suppression methodology in

Xenopus oocytes, it is worth mentioning other methods for incorporating unnatural amino

acids into proteins. The auxotrophic strain method of mutagenesis of Tirrell and co-

workers and the evolution of orthogonal tRNA/synthetase pairs of Schultz and co-

workers are described in detail in recent review articles19-21 and are briefly discussed

below. The auxotrophic strain method allows for residue-specific replacement of amino

acids with unnatural residues in E. coli. By depleting an auxotrophic bacterial host of the

natural amino acid and supplementing the system with the unnatural analogue, the Tirrell

9

method enables chemical modification of proteins at multiple sites.20 This method is able

to produce significantly more protein than the above-mentioned in vivo nonsense

suppression methodology, but is limited to studies where residue-specific mutations are

desired.

Recent advances for the site-specific introduction of unnatural amino acids by Schultz

and co-workers significantly improve protein expression efficiency in comparison to in

vivo nonsense suppression. In this method, the evolution of an orthogonal aminoacyl-

tRNA synthetase to misaminoacylate a suppressor tRNA with an unnatural amino acid in

E. coli, results in high protein yields.19,21 This method has not been optimized for

residues such as α-hydroxy acids that are metabolized in the cell and therefore are unable

to acylate the suppressor tRNA in vivo.21 Recently, Dougherty,22 Schultz,21 and Sisido23

achieved site-specific incorporation of two different unnatural amino acids into the same

protein by utilizing two different quadruplet stop codons. These methods provide

powerful tools for conducting structure-function studies of proteins and potentially for

creating proteins with enhanced therapeutic properties.

1.3 NICOTINIC ACETYLCHOLINE RECEPTORS

Nicotinic acetylcholine receptors (nAChR) are the most extensively studied members of

the Cys-loop family of LGICs. The embryonic mouse muscle nAChR is a

transmembrane protein composed of five subunits, (α1)2β1γδ (Figure 1.5). Each subunit

contains an extracellular ligand-binding domain at the N-terminus and four

transmembrane domains (TM1-4). The second transmembrane domain, TM2, lines the

interior of the channel pore. Early biochemical studies identified two agonist-binding

10

sites at the α/γ and α/δ interfaces on the muscle type nAChR that are defined by an

aromatic box of conserved amino acid residues.24,25 The principal face of the agonist-

binding site contains four of the five conserved aromatic box residues, while the

complementary face contains the remaining aromatic residue.

Figure 1.5. nAChR Subunit Arrangement.26 Overall layout of the mouse muscle nAChR showing (α1)2β1γδ subunits. The bindings sites reside at the interface of α/γ and α/δ subunits, where the majority of the binding site residues reside on the primary α subunit. Figure adapted from reference 26 by permission from Macmillan Publishers Ltd: Nature, copyright (2001).

In the past 5 years, knowledge of the ligand-binding domain has been dramatically

advanced by the discovery27 and crystallization of the acetylcholine-binding protein

(AChBP).5 AChBP is a homopentamer isolated from the snail Lymnaea stagnalis and it

shares approximately 20 % sequence homology with the nAChR extracellular ligand-

binding domain. In 2001, a 2.7 Å crystal structure of the acetylcholine-binding protein

(AChBP)5 confirmed early biochemical studies and provided additional structural

information on the ligand-binding domain (Figure 1.6 A, B). Sixma and co-workers also

11

published a 2.2 Å nicotine-bound AChBP structure and a 2.5 Å carbamylcholine-bound

AChBP structure in 2004.28 In addition, the crystal structures of AChBP were solved in

2005 from the Bulinus truncatus29 and Aplysia californica snail species.30 These

structures greatly impacted the field by providing insight for studies examining ligand-

receptor interactions and by aiding in drug discovery.

While high resolution crystallographic data is difficult to obtain for transmembrane

nAChRs, studies by Unwin and co-workers shed light onto the structure of the full-length

receptor.6,31-33 The structure of full-length nAChR was determined at 9 Å 31 and later at

4.6 Å resolution using cryo-electron microscopy.32 More recently, Unwin generated a

refined model at 4.0 Å resolution of Torpedo nAChR using insights from the AChBP

structures (Figure 1.6 C, D).

These structures represent only static pictures and do not provide information on how

these dynamic proteins transition from one conformation to another. Thus, more

knowledge of protein transitions on the atomic level is still needed to fully understand

structural rearrangements that occur when ligand binding at the agonist-binding site leads

to the gating of channel residues nearly 60 Å away. The structural models of the full-

length nAChR provide insights for recent studies that make significant advances towards

understanding the gating mechanisms of these membrane proteins.13,34,35 The complete

mechanism, however, that couples ligand binding to channel gating of Cys-loop receptors

is not fully understood and remains an important topic in molecular neuroscience.

12

Figure 1.6. Structural Information for AChBP and nAChR. A) AChBP structure with two subunits highlighted and the binding site residues indicated.5 Figure reprinted from reference 5 by permission from Macmillan Publishers Ltd: Nature, copyright (2001). B) The aromatic binding site of AChBP with muscle nAChR numbering. Black numbers indicate α subunit residues while blue numbers indicate non-α subunit residues. C) Top view of Unwin’s refined 4 Å cryo-electron model.6 D) Side view of Unwin’s 4 Å model indicating the extracellular, transmembrane, and intracellular regions.6 Figures C & D reprinted from reference 6 with permission from Elsevier.

13

1.4 nAChR DRUG-RECEPTOR INTERACTIONS

The primary focus of the present work is to gain an understanding of ligand-receptor

interactions at the mouse muscle nAChR. We utilize chemical scale investigations to

identify mechanistically significant drug-receptor interactions at the muscle-type nAChR

as predicted by AChBP structures. Interestingly, structurally similar nAChR agonists

acetylcholine (ACh), nicotine, and epibatidine (Figure 1.7) are known to bind to the

same region of the protein while eliciting different responses in protein activity. These

three agonists also display different relative activity among different nAChR subtypes. A

better understanding of residues that play a role in determining agonist activity and

specificity would provide insight into mechanisms that underlie agonist binding and

channel gating. This information could also aid in designing nAChR therapeutics.

Acetylcholine Nicotine Epibatidine

Figure 1.7. Structures of nAChR Agonists.

The goals of this thesis are threefold. First, the study incorporates unnatural amino

acids at the ligand binding site to probe agonist binding determinants that differentiate

acetylcholine, nicotine, and epibatidine agonist nAChR activity. Second, the study

identifies residues in the shell of amino acids immediately surrounding the agonist

binding box that are important in shaping the ligand binding site for all three agonists.

+ +

14

Third, the study examines residues surrounding the agonist-binding site that contribute to

ACh, epibatidine, and nicotine specificity.

1.5 DISSERTATION SUMMARY

The work presented in this thesis centers on drug-receptor interactions at the mouse

muscle nAChR. Chapter 2 describe studies that probe the binding of three distinct

agonists–acetylcholine (ACh), nicotine, and epibatidine–to the nAChR using unnatural

amino acid mutagenesis. Results from these studies reveal how three structurally similar

agonists bind to the same binding site through quite different non-covalent binding

interactions to activate the receptor. This chapter is based on a Journal of the American

Chemical Society paper written in collaboration with E. James Petersson. James

Petersson conducted computational modeling studies to supplement the experimental

data.

The work in Chapter 3 describes the ability of conserved residues immediately

outside of the aromatic binding box to interact with binding site residues and to play a

role in determining nAChR activity. Part A of this work examines a network of hydrogen

bonds between an outer shell residue and residues in the aromatic box. These studies

demonstrate an important role for this residue in stabilizing the agonist-binding site.

These studies were performed in collaboration with Michael Torrice who designed an

unnatural amino acid to probe the importance of charge in this region. Part B of this

work examines a highly conserved residue immediately surrounding the agonist binding

box that is proposed to reposition its side chain upon ligand binding. With additional

15

evidence from other recent advances, this site is proposed to be important in initiating the

nAChR channel gating pathway.

The work in Chapter 4 utilizes computational protein design to probe residue

positions that affect nAChR agonist specificity for acetylcholine, nicotine, and

epibatidine. Results from these studies identify mutations that enhance nAChR

specificity for nicotine, over ACh and epibatidine compared to wild-type receptors. This

project was conceptualized through collaboration with Jessica Mao in Steve Mayo’s lab

who generated the computational predictions.

Finally, Chapter 5 reflects studies conducted prior to candidacy in Peter Dervan’s lab.

Part A of this work evaluates the ability of a series of cationic polyamides to enhance

polyamide affinity while maintaining specificity by varying the number, relative spacing,

and linker length of aminoalkyl side chains. These studies were performed in

collaboration with Ben Edelson who synthesized N-aminohexyl and N-aminodecyl

pyrrole containing polyamides. Part B of this work examines the nuclear uptake

properties of these polyamides in mammalian cells, also performed in collaboration with

Ben Edelson.

16

1.6 REFERENCES

1. Paterson, D. & Nordberg, A. Neuronal nicotinic receptors in the human brain. Progress in Neurobiology 61, 75-111 (2000).

2. Cassels, B. K., Bermudez, I., Dajas, F., Abin-Carriquiry, J. A. & Wonnacott, S. From ligand design to therapeutic efficacy: the challenge for nicotinic receptor research. Drug Discovery Today, 1657-1665.

3. Long, S. B., Campbell, E. B. & MacKinnon, R. Crystal structure of a mammalian voltage-dependent Shaker family K+ channel. Science 309, 897-903 (2005).

4. Celie, P. H. N. et al. Nicotine and Carbamylcholine Binding to Nicotinic Acetylcholine Receptors as Studied in AChBP Crystal Structures. Neuron 41, 907-914 (2004).

5. Brejc, K. et al. Crystal structure of an ACh-binding protein reveals the ligand-binding domain of nicotinic receptors. Nature 411, 269-76 (2001).

6. Unwin, N. Refined structure of the nicotinic acetylcholine receptor at 4A resolution. Journal of Molecular Biology 346, 967-89 (2005).

7. Bass, R. B., Strop, P., Barclay, M. & Rees, D. C. Crystal structure of Escherichia coli MscS, a voltage-modulated and mechanosensitive channel. Science 298, 1582-7 (2002).

8. Jiang, Y. X. et al. X-ray structure of a voltage-dependent K+ channel. Nature 423, 33-41 (2003).

9. Coe, J. W. et al. Varenicline: an alpha4beta2 nicotinic receptor partial agonist for smoking cessation. Journal of Medicinal Chemistry 48, 3474-7 (2005).

10. Nowak, M. W. et al. In vivo incorporation of unnatural amino acids into ion channels in Xenopus oocyte expression system. Ion Channels, Pt B 293, 504-529 (1998).

11. Beene, D. L., Dougherty, D. A. & Lester, H. A. Unnatural amino acid mutagenesis in mapping ion channel function. Current Opinion in Neurobiology 13, 264-70 (2003).

12. Tong, Y. H. et al. Tyrosine decaging leads to substantial membrane trafficking during modulation of an inward rectifier potassium channel. Journal of General Physiology 117, 103-118 (2001).

13. Lummis, S. C. et al. Cis-trans isomerization at a proline opens the pore of a neurotransmitter-gated ion channel. Nature 438, 248-52 (2005).

14. Lummis, S. C., Beene, D. L., Harrison, N. J., Lester, H. A. & Dougherty, D. A. A cation-pi binding interaction with a tyrosine in the binding site of the GABAC receptor. Chemistry & Biology 12, 993-7 (2005).

15. Monahan, S. L., Lester, H. A. & Dougherty, D. A. Site-specific incorporation of unnatural amino acids into receptors expressed in mammalian cells. Chemistry & Biology 10, 573-580 (2003).

16. England, P. M., Lester, H. A. & Dougherty, D. A. Incorporation of esters into proteins: Improved synthesis of hydroxyacyl tRNAs. Tetrahedron Letters 40, 6189-6192 (1999).

17. Petersson, E. J. in Chemistry (Caltech, Pasadena, 2005).

17

18. Noren, C. J., Anthonycahill, S. J., Griffith, M. C. & Schultz, P. G. A General-Method for Site-Specific Incorporation of Unnatural Amino-Acids into Proteins. Science 244, 182-188 (1989).

19. Wang, L. & Schultz, P. G. Expanding the genetic code. Angewandte Chemie-International Edition 44, 34-66 (2005).

20. Link, A. J., Mock, M. L. & Tirrell, D. A. Non-canonical amino acids in protein engineering. Current Opinion in Biotechnology 14, 603-609 (2003).

21. Xie, J. M. & Schultz, P. G. Adding amino acids to the genetic repertoire. Current Opinion in Chemical Biology 9, 548-554 (2005).

22. Rodriguez, E. A., Lester, H.A. & Dougherty, D. A. In Vivo Incorporation of Multiple Unnatural Amino Acids Using Nonsense and Frameshift Suppression. Submitted (2005).

23. Hohsaka, T., Ashizuka, Y., Taira, H., Murakami, H. & Sisido, M. Incorporation of nonnatural amino acids into proteins by using various four-base codons in an Escherichia coli in vitro translation system. Biochemistry 40, 11060-4 (2001).

24. Grutter, T. & Changeux, J. P. Nicotinic receptors in wonderland. Trends in Biochemical Sciences 26, 459-463 (2001).

25. Karlin, A. Emerging structure of the nicotinic acetylcholine receptors. Nature Reviews Neuroscience 3, 102-14 (2002).

26. Dougherty, D. A. & Lester, H. A. Neurobiology - Snails, synapses and smokers. Nature 411, 252-254 (2001).

27. Smit, A. B. et al. A glia-derived acetylcholine-binding protein that modulates synaptic transmission. Nature 411, 261-8 (2001).

28. Celie, P. H. et al. Nicotine and carbamylcholine binding to nicotinic acetylcholine receptors as studied in AChBP crystal structures. Neuron 41, 907-14 (2004).

29. Celie, P. H. et al. Crystal structure of acetylcholine-binding protein from Bulinus truncatus reveals the conserved structural scaffold and sites of variation in nicotinic acetylcholine receptors. Journal of Biological Chemistry 280, 26457-66 (2005).

30. Hansen, S. B. et al. Structures of Aplysia AChBP complexes with nicotinic agonists and antagonists reveal distinctive binding interfaces and conformations. Embo Journal 24, 3635-46 (2005).

31. Unwin, N. Nicotinic acetylcholine receptor at 9 A resolution. Journal of Molecular Biology 229, 1101-24 (1993).

32. Miyazawa, A., Fujiyoshi, Y., Stowell, M. & Unwin, N. Nicotinic acetylcholine receptor at 4.6 A resolution: transverse tunnels in the channel wall. Journal of Molecular Biology 288, 765-86 (1999).

33. Miyazawa, A., Fujiyoshi, Y. & Unwin, N. Structure and gating mechanism of the acetylcholine receptor pore. Nature 423, 949-55 (2003).

34. Xiu, X., Hanek, A. P., Wang, J., Lester, H. A. & Dougherty, D. A. A unified view of the role of electrostatic interactions in modulating the gating of Cys loop receptors. Journal of Biological Chemistry 280, 41655-66 (2005).

35. Grosman, C., Zhou, M. & Auerbach, A. Mapping the conformational wave of acetylcholine receptor channel gating. Nature 403, 773-6 (2000).